# Release Readiness

> Assess release evidence without publishing, tagging, or changing remote systems.

- Skill: `prashantyr/release-readiness` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add prashantyr/release-readiness`
- Raw SKILL.md: https://api.skillmd.com/api/skills/prashantyr/release-readiness/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: prashantyr (https://skillmd.com/u/prashantyr)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/prashantyr/release-readiness

---


# Release Readiness

Use before a software release or deployment handoff.

1. Identify version, target, rollback owner, risk tier, and required approvals.
2. Run the repository's tests, lint, build, and security checks. Capture exact commands/results.
3. Run `python release-readiness/scripts/check.py <directory>` to check baseline artifacts.
4. Confirm version/changelog consistency, migration reversibility, artifact provenance, dependency
   review, observability, staged rollout, rollback instructions, and unresolved high-severity issues.
5. Return `ready`, `conditionally ready`, or `not ready`, with evidence and explicit blockers.

The script is read-only. Never tag, push, publish packages/images, approve a release, change CI
secrets, or deploy unless separately and explicitly authorized. Prefer local tools and Ollama for
summaries. Commercial providers are optional and must receive redacted data through an explicit
adapter using environment-managed credentials.